After closely reading the letter by Carlosama-Rosero in which we were cited,1 we observed some misinterpretations of the content of our publication2 and we believe it would be useful to provide some clarifications.

In our previous letter, we reference the mental process known as argumentum ad ignorantium, which consists of believing that lack of awareness of evidence contrary to an idea that we support counts as evidence in its favor. At that time, we gave the example of the drug ivermectin in COVID-19 disease, whose previous success in a laboratory study would be difficult to extrapolate to clinical practice, as it would require toxic doses to reach the necessary therapeutic concentration.2

Although it is well-known that ivermectin dosing is in µg/kg of body weight, we gave the total necessary dosage in mg for reaching a concentration in human beings that is similar to what was effective in vitro, which would indeed be highly toxic (1000–1200 mg vs 10–20 mg, which is the dose habitually used in humans). At no point did we intend to categorically rule out ivermectin as a possible treatment for COVID-19, but rather refute inconsistent evidence in its favor. The work by Rajter et al.3 that Carlosama-Rosero cites is a multicenter retrospective study of 280 patients treated with ivermectin which reported a significant reduction in the overall mortality rate in the group that received ivermectin (15% vs 25.2%; OR 0.52 [CI 95% 0.29–0.96], p = 0.03). Furthermore, on the regression analysis adjusted for confounding variables, the reduction in mortality remained significant (OR 0.27 [CI 95% 0.09–0.85], p = 0.03; HR 0.37 [CI 95% 0.19–0.71], p = 0.03). However, like all observational studies, it had some well-known selection and confounding biases, and therefore the results reported in that publication must be taken with great caution and of course after waiting for the results of ongoing clinical trials. This work unfortunately does not allow for drawing solid conclusions on the effectiveness of ivermectin at this time.